284 Effect of Cannon in dissipating 



kept up a continual fire which overthrew a part of the entrench- 

 ments of the town ; on the approach of night, upon a signal 

 given by the commandant, there followed a general firing from 

 the batteries and vessels, and notwithstanding, all this did not hin- 

 der the onset of a storm, accompanied, remarks the admiral, with 

 redoubled claps (rf'fearfid thunder and lightnings zvhich suc- 

 ceeded each other with scarcely any interval. Here, then, we 

 have an eocperiment^ in which are, assuredly, to be found all the 

 favourable conditions possible, and notwithstanding, these many 

 thousand detonations, infinitely more intense than those of the 

 small cannon and mortars of the Maconnais, did not prevent 

 the thunder-storm Jromjormmgi and when it was formed, they 

 did Qiot disperse it. 



If a single fact^ that borrowed from Forbin, has not, as it 

 would appear, satisfactorily demonstrated that the reports of 

 artillery have not the power of dissipating thunder-storms, it is 

 as clear on the other hand, that the isolated fact ^ derived from 

 the memoirs of Duguay-Trouin, does not prove the opposite 

 position. It cannot, however, be questioned that those who have 

 at their command the detailed annals of the late wars, will there 

 find a multitude of documents more than sufficient to elucidate 

 the question before us. I shall here repeat two facts which oc- 

 cur to my own memory, in the hope that they will lead to ana- 

 logous statements. The 25th of August 1806, being the day 

 selected for the attack of the islet and fortress of Dannholm, 

 near Stralsund, General Fririon, that he might harass and fa- 

 tigue the Swedish garrison, ordered it to be cannonaded du- 

 ring the whole day. In spite of these powerful and continued 

 discharges of artillery, a violent thunder-storm visited the spot 

 at nine o"*clock in the evening. Again, it happened oddly 

 enough, that the English line-of-battle-ship, the Dul^e^ of 90 

 guns, was struck with lightning in the year 1793, whilst it was 

 cannonading one of the batteries of Martinico. 



Finally, I may here adduce another observation which, in 

 default of more direct experiments, may not prove uninteresting. 

 In the wood of Vincennes, at the distance of nearly two leagues 

 from the observatory of Paris, there is a polygon fort where the 

 artillery are in the habit of practising during certain months of 

 the year. This fort is supplied with eight battering guns 



